The middle school years are peak years for initiation of alcohol and marijuana (Johnston et al., 2004). Unfortunately, most youth who engage in substance use and experience problems are unlikely to voluntarily make use of formal prevention services (D'Amico, 2005;Johnson et al., 2001;Wu et al., 2003). A small body of recent research suggests that youth may benefit from less formal programs that are brief, voluntary, and easily accessible (Brown, 2001;Brown et al., 2005;D'Amico et al., 2004;D'Amico &Orlando, 2005). However, very few intervention programs of this type have been developed (Little and Harris, 2003). Thus, while this approach shows promise, the impact of intervention programs that younger teens may choose to attend has not been extensively examined. One such intervention, Project CHOICE, was developed and tested by the PI using NIAAA funding for developmental work (R21AA13284-01). Project CHOICE is the only voluntary intervention that has been tested for middle school youth and our small quasi-experimental study has demonstrated its efficacy in one school setting (D'Amico et al., 2005;D'Amico &Orlando, 2005). The Project CHOICE intervention addresses several critical gaps in the field, including beginning to understand voluntary service utilization among this age group and assessing how this type of program may impact school-wide use of alcohol and other drugs (AOD). The main objective of the proposed 5-year longitudinal study is to build on our initial work by conducting a more rigorous test of Project CHOICE. The study will include 16 middle schools, located in the ethnically diverse Southern California cities of Los Angeles, Santa Monica, and Torrance. These schools will be randomly assigned as intervention (n= 8) or control (n = 8). We will first examine individual-level effects by testing whether Project CHOICE affects AOD-related outcomes among students who participate in the intervention. We will then examine school-level effects by testing whether AOD rates among all students in the intervention schools are affected, regardless of participation. We assume that these school-level effects will be due to changes in the school environment (e.g., Project CHOICE advertising, discussion of Project CHOICE among students, changes in social norms). In anticipation of this school-level impact, a secondary objective of this study is to gain a better understanding of who participates in Project CHOICE, as well as how these participants and changes in the school environment may influence the attitudes and behaviors of those who do not participate. This research incorporates a novel methodology for AOD involvement, as it emphasizes personal self-change efforts and natural recovery and is appealing to both using and non-using youth. The work proposed in this application represents the important next step in this line of research: to more critically evaluate Project CHOICE and its potential impact on both school-wide and individual outcomes with a larger population of youth.